'But For' Test

Image: TanteTati/Pixabay

• Imaginary Encounter on the Clapham Omnibus

Oana, a seven-year-old from a nearby farmstead, cuts off the string of a butterfly-shaped kite at one bank of a river in the hope that the kite will reach her mother working in the town on the other bank. As the kite drifts across a field on the other bank, large eyespots painted on it startle a baby in a pram, who then wails incessantly. Its parents quarrel all night long over how they should have placated it. Still in a foul mood from the fight the next morning, the baby's father announces salary cuts and extra overtime at his factory. One floor leader, feeling smaller and more inadequate than ever, drives to a car wash after lunch, where he unleashes a stream of expletives at an attendant for her supposed snail pace. Fuming more each time she replays the encounter in her head, the attendant shoves and elbows commuters in the way on her subway trip home. Among them is one hooded teenager whose remaining shred of optimism about humanity is now shoved aside. A homemade bomb explodes at a station the next day. Under police interrogation, each traceable member of the chain blames the person upstream. And so, if we would believe it, a pint-sized girl playing with a kite has killed four and injured dozens twenty miles away.

• What is the 'But For' Test?

Generally used as one of multiple steps in determining a defendant's responsibility for an accused wrongdoing in tort law (a body of law concerned with members of a community suing fellow members for causing various types of non-contractual harms) and criminal law (states suing harm-inducing members on communities' behalf), the test asks this question:

Would the harm have visited upon the victim if the defendant had not performed or omitted an act?*1*2

• How Can the 'But For' Test be Conducted?

One conceivable way of perceiving the 'but for' test is this: Instead of obsessively hunting down the ultimate point of genesis of a chain of events, it examines whether the chain could have stopped or changed path at a certain stage, before the final outcome emerged. In the chain fingerpointing example, the issue for the hooded teenager is whether casualties from the explosion would have resulted had he not planted the bomb. The answer is overwhelmingly likely negative. Car Wash Attendant's behaviour was understandably galling and, as was the conduct of most of those further upstream, morally problematic. Still, it is not inevitable that one person's brusqueness would cause another person's act of mass murder. Hooded Teenager had a choice to take the chain of events in more than one direction: forbearance, verbal protest, letter to the local newspaper about public transport incivility, or, far more undesirably, online shaming, to name only a few physically non-injurious possibilities. 

Someone in Hooded Teenager's shoes might complain that it is unjust to analyse whether the chain could have been interrupted at his stage without considering the possibilities of its interruption at other stages. Indeed, if anyone earlier in the chain had reined in their inconsiderate behaviour, this explosion might not have occurred. Maybe he would just reach home uneventfully and receive counselling before any other person triggers an outburst from him. Why is the 'but for' test not applied to them? 

That disparity may be understood by seeing parties without the requisite duty of care to would-be victims as exempt from such questioning. In the absence of such a limit, the scope of queries might be so wide—'What if Hooded Teenager's parents had not conceived him?' 'What if his grandparents had not married?' 'What if the government had provided more rural work opportunities for Oana's absent mother?'—as to render everyone culpable for everything under the Sun.

A natural issue of concern at this juncture is what should determine the existence of the above duty. Hooded Teenager's advocates might retort that the possibility of being agitated into physical violence, while not inevitable, was still among the possible directions of the chain of events and thus warranted Car Wash Attendant's care. There are a few problems with this defence. To begin with, given how even the minutest of affairs can spark violent impulses in the darkest of minds, such a possibility is present with every action or even inaction, the latter of which can imaginably be interpreted as patronisation, social exclusion, camouflaging of mass conspiracy, etc. The prevalence of the possibility renders it so impractical to forestall altogether that its mere presence is meaningless to decisionmaking in everyday life. The more useful consideration is the likelihood of the possibility. Here, we could borrow the reasonable person standard and test of foreseeability commonly employed in court rulings to supplement the 'but for' test: Would a reasonable individual have reacted to a shove with mass killings? Could Car Wash Attendant have reasonably foreseen this reaction when most people do not venture anything close? The related test of remoteness*3 could be of assistance too, more so if we were questioning the liability of people earlier in the chain: Were their acts too far removed from the psychological impact on the teenager? Were there too many sequential links between their acts and this psychological impact?

The argument for Car Wash Attendant's blameworthiness does receive a potential boost if there were markings on the teenager that hinted at his volatility (i.e. he probably should not be treated as a reasonable agent). Perhaps, he was visibly emotionally tortured. Perhaps, he was carrying a large bomb-making manual. Perhaps, his sweater explicitly read, 'If someone pushes me, I'll set off a bomb in public tomorrow.' Nonetheless, it must be noted that multiple parties can hold responsibility for the same incidence of harm. The contention that someone else is at fault, even if true, would not automatically absolve one of blame. Accessories to crimes (i.e. people who assists or encourages the perpetration of crimes, generally without appearing at the crime scenes) are punished along with, not in place of, autonomous criminals of sound mind performing the final legwork. While an insanity defence is indeed available for an infirm person who meets the criterion, provocation serves as a mitigating factor at best, not as a defence that fully absolves an accused of guilt in court.*4

• Strengths of the 'But For' Test

1. Even outside jurisprudential contexts, the test throws into relief personal autonomy, counteracting any tendency to shift blame on the part of the individual undergoing the test. It can be a reminder to litigants and observers alike that history and circumstances need not determine anyone.
2. For someone whose conduct is under examination, the contrarian tone of the phrase 'but for' encourages exploration beyond one's own instincts, dogmatic thinking and self-centred interests. That creates room for standing in adversaries' and victims' shoes.*5
3. The concept is simple to understand. Ceteris paribus (a beautiful phrase for the pristine, ideal state where all other things are equal), this is potentially a boon for widespread usage, lesson absorption and speedy problem-solving. Compared to some umbrella theory that strives to capture all possible patterns of harm causation, its bright-line approach facilitates likely faster processing of a set of relatively straightforward cases.

• Vulnerable Spots

1. In narrowing the application of the 'but for' test to those with the requisite duty of care, there is the somewhat paradoxical and perverse risk of shifting the general duty to treat everyone one comes into contact with kindly down the chain, to the juncture where concrete harm occurs, or at least creating an impression of doing so. Law does not police such a duty, so it is the implementation of the test in non-legal settings that is of concern. Under communication climates where emotional inflation, short attention spans, snap judgments and association of like with like are the norm, the manner in which the test is carried out may lead to warped conclusions. These include: (i) because Hooded Teenager is at fault for the visible damage, everyone else in the chain in such situations is faultless in all respects and does not ever need to concern themselves with potential emotional snowballing; and (ii) society, in pinning the blame for such damage squarely on Hooded Teenager and the like, slyly, self-indulgently and unfeelingly exculpates all other parties of fault for crude, mean and unthoughtful behaviour. 
2. The reasonable person standard, test of foreseeability and test of remoteness are vague in many instances. Within a community, opinions often diverge as to what ordinary sensibility should entail. Neither is it always clear what kinds of factors warrant precaution, what the probability of a risk materialising is, what probabilities merit a different behaviour, and when a possible consequence is so indirect and implicates parties so beyond the blamed party's circle of association that it need not oblige consideration beforehand. Is a reasonable person supposed to assume that someone carrying a bomb-making manual intends to make a bomb, and for homicidal purposes at that? A noir writer of any age may have reason to lug such references around too. And would Car Wash Attendant be unreasonable to believe any warning on Hooded Teenager's sweater to be a hyperbolic expression of adolescent angst no more serious than death threats parents use to elicit obedience from children or part of some incomprehensible youth sarcasm subculture? 'Common' sense*6 can be elusive even when a risk is quantifiably small. In Chester v Afshar, it took the UK House of Lords a narrow 3:2 majority to hold a doctor liable for not informing a patient of a surgical risk, which probability of occurrence was a mere 1-2%.*7
3. Assessments probably carry inaccuracies. 'But for' propositions rely on counterfactuals*8—what would have transpired in alternate timelines in which the blamed party behaved differently—yet there is very frequently no way to divine such outcomes with perfect certainty. The result is that test executors may end up ignoring the potential of the human will and relegating victims of alleged wrongdoings to ordinary forces of past circumstances. A fictional example would be a bereaved family blaming a wish-fulfilling charity for reneging on its promise to arrange for their then-terminally ill son to host a current affairs show he liked, only for the charity to counter that, with his tender age and very severe stuttering condition, no broadcaster and audience could have accepted his hosting in a real sense. While that retort would sound reasonable if all the child wanted to be was a proper, successful host and this precise failure was the subject of the blame, it discounts any possibility of him winning over prospective bosses and viewers with charm, precocity, unexpected professionalism in other aspects and sheer grit. Nonetheless, a line may need to be drawn somewhere in many cases because, as with the consideration for duty of care, to entertain any kind of cause-effect possibility is to make way for impossibility of decisionmaking.
4. As a solution, the test is simplistic. The risk of censuring individuals for flouting duties of care in situations beyond their control necessitates considerations of intention and duress, on top of those of reasonableness, foresight and mental capacity alluded to above, although this is not always seen as strategically ideal in legal contexts. In a perfect world, we might also want people to be otherwise held responsible for every act of endangerment, whether or not harm is fortuitously averted or so mixed up with other proximate factors (e.g. existing injury; multiple parties engaged in the same, wrongful behaviour; freak disaster subsequent to wrongdoing) that it might/would have occurred anyway. Court-developed law, which the 'but for' test is usually*9 an instance of, on the other hand, is more concerned with actual detriment.*10 Nevertheless, legal scholars and decisionmakers have come up with various complementary and alternative means of establishing fault, even if policy and morality remain unidentical matters. Prominent ones include: test for material contribution to harm*11 or risk of harm,*12 recognition of lost chance larger than 50%,*13 sufficient concurrent cause standard (i.e. check for whether the conduct is enough to cause the harm by itself),*14 NESS (Necessary Element of a
   Sufficient Set) test,*15 acceleration theory (i.e. positing the 'but for' query for the harm at its time of occurrence, not just any possible time)*16,*17 and sanctions for attempts at crimes.*18
5. Taken superfically, the test might conflate causation with correlation.*19 Conceivably, there can be situations where harm would consistently visit upon Person A if Person B behaves in a certain manner, even though this pattern is (i) a peculiar string of coincidences, (ii) a pair of outcomes from a common cause (e.g. Person C, Dad, hitting Person B, Mom, before hitting Person A, the young child annoying him with her cries after witnessing the attack—'But for Mom shrieking whenever she receives a beating, Girl would not have been hit; therefore Mom is responsible for Girl's domestic abuse!'), or (iii) a result of A harming B (e.g. B kicks out his leg as A shrieks, leading to the implicative observation 'But for B's kicking, A would not have shrieked', even though A is in reality shrieking in mockery as he hits B in the knees, perhaps from an angle obscured to the viewer, and B's leg gives a reflex response). 

• Food for Thought 

Those to whom evil is done

Do evil in return.

—W. H. Auden's 'September 1, 1939'

It is erroneous to assume that any negative energy you pass on to a party will stay or dissipate within her. The urge to reclaim power and correct some perceived moral justice calculus very frequently prevails. All the same, law exists not only to care for our natural desires and vulnerabilities but also to stimulate us to counter them during the times they threaten destruction. It can and ought to be an encouragement to (i) confront those who do harm to us directly, instead of cowardly unleashing anger at those weaker than us at the opportunity to do so; and (ii) tap into the creative and rational sides of the human mind, not its primitive core, so that non-destructive, perhaps even win-win solutions may be found. Point to a positive trait (e.g. an immaculate outfit suggesting high standards for oneself, vegan backpack, or old watch of possibly sentimental value) of the offensive person, for instance, and state firmly that the wrongful behaviour is creating an unfairly lowly impression of him or her. Ask if he or she is all right, recommending a rest before finalising the decision or completing the service/journey.

If that does not work or is too dangerous to attempt, head for due process. Report the incident to the police / security department / disciplinary board / worker welfare agency, etc. Spread awareness about the need for more considerate behaviour. Rally for legal reforms and tougher law enforcement that would create an environment where one can confront a perpetrator of sound mind without sacrificing personal safety. Protect others from also falling victim. We often wonder about the meaning of existence, yet there are so many improvement tasks around we see so much meaning in with so huge a shortage of takers.

Even if changes and apologies seem to be a long way away, take immense pride in the knowledge that we are still the bigger people. Revenge might be sweet but buying into the same perverse logic and measuring ourselves by the same crude standards as our adversaries would make us no different from them. Play your game, not theirs!

• More Servings

  www.bloomsburyprofessional.com

  
  

*1:Kirsty Horsey and Erika Rackley, Tort Law (5th edn, Oxford University Press 2017) 248-49.

*2:Jonathan Herring, Criminal Law (7th edn, Palgrave Macmillan 2011) 47 and 55.

*3:For a review of the test, see Victor E Schwartz, 'Remoteness Doctrine: A Rational Limit on Tort Law' (1999) 8 Cornell JL & Pub Pol'y 421.

*4:Gerry Ferguson, 'Criminal Liability and Criminal Defenses' in James D Wright (ed), International Encyclopedia of the Social & Behavioral Sciences (2nd edn, Elsevier 2015).

*5:Robert N Strassfeld, 'If . . .: Counterfactuals in the Law' (1992) 60 Geo Wash L Rev 339, 415.

*6:Mayo Moran, Rethinking the Reasonable Person: An Egalitarian Reconstruction of the Objective Standard (Oxford University Press 2003) 157(describing and criticising invocations of common sense in legal rhetoric surrounding the reasonable person standard).

*7:[2004] UKHL 41, [2005] 1 AC 134.

*8:Helmut Weber, 'The 'But For' Test and Other Devices—The Role of Hypothetical Events in the Law' (2009) 34 Hist Soc Res 118.

*9:For an occurrence of the test as a statutory rule, see Criminal Code Act 1924 (Tas), s 153(2). This is criticised in David Lanham and others, Criminal Laws in Australia (The Federation Press 2006) 156.

*10:R (on the application of Burke) v General Medical Council [2005] EWCA Civ 1003, [2006] QB 273 [21]-[22] ('There are great dangers in a court grappling with issues [...] divorced from a factual context that requires their determination.'). All the same, judicial decisions often take into account potential preemptive effects.

*11:e.g. Wardlaw v Bonnington Castings Ltd [1956] UKHL 1, [1956] AC 613.

*12:e.g. Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, [2003] 1 AC 32.

*13:e.g. Allied Maples Group Ltd v Simmons & Simmons (a firm) [1995] EWCA Civ 17, [1995] 1 WLR 1602.

*14:Paul H Robinson and Tyler Scot Williams, Mapping American Criminal Law: Variations Across the 50 States (ABC-CLIO 2018) 67.

*15:Richard W Wright, 'Causation, Responsibility, Risk, Probability, Naked Statistics, and Proof: Pruning the Bramble Bush by Clarifying the Concepts' (1988) 73 Iowa L Rev 1001, 1019-21.

*16:Laurie L Levenson, The Glannon Guide to Criminal Law (5th edn, Wolters Kluwer 2018) 129.

*17:e.g. Oxendine v State, 335 P2d 940, 1958 OK CR 104 (Okla Crim App 1958).

*18:e.g. Criminal Attempts Act 1981 (UK); State v Mitchell, 170 Mo 633, 71 SW 175 (1902).

*19:Weber, 122-23.